THeres a chunk where i created a result column in the other markdown above the chunk below. Not sure if I did it again later on so I’m not including that for now

Horizontal break is meaningless if theres no distinction made between lefties and righties

pitches$Hbrk <- NA

pitches$Hbrk[pitches$PitcherThrows == 'Right'] <- -pitches$HorzBreak

## Warning in pitches$Hbrk[pitches$PitcherThrows == "Right"] <- -pitches$HorzBreak:
## number of items to replace is not a multiple of replacement length

pitches$Hbrk[pitches$PitcherThrows == 'Left'] <- pitches$HorzBreak

## Warning in pitches$Hbrk[pitches$PitcherThrows == "Left"] <- pitches$HorzBreak:
## number of items to replace is not a multiple of replacement length

pitches %>% group_by(PitcherThrows) %>% summarise(n =n())

##Year Averages used for differences

##Find averages for the dependent measures; Velovity, Spin Rate, Break,     Still try to do command, result based metric


pitcher_year_group <- sqldf("select PitcherId, TaggedPitchType, SUBSTRING(Date, 1, 4) as YEAR, MAX(Pitcher) as PitcherName, MAX(PitcherTeam) as Team, AVG(RelSpeed) AS avg_RelSpeedYear, AVG(SpinRate) AS avg_spinrateYear, AVG(InducedVertBReak) AS avg_IndVertBrkYear, AVG(Hbrk) As avg_HBrkYear, COUNT(*) as PitchesThrown FROM pitches  WHERE PitcherId is not null GROUP BY PitcherId, SUBSTRING(Date, 1, 4), TaggedPitchType ORDER BY COUNT(*) DESC")

Relevel Pitch_group. MAke 0-9 the baseline

pitches$pitch_group <- relevel(factor(pitches$pitch_group), ref = '0-9 Pitches')

Next we have to merge it so that summary stats for each pitch type of each pitcher are included for in each row, later we’ll also calculate differences between each individual pitch numbers and the averages

pitches$YEAR = NA
pitches$YEAR <- substr(pitches$Date, 0, 4)
pitcherYearAvgs <- left_join(pitches, pitcher_year_group, on = c('PitcherId', 'YEAR'))

## Joining, by = c("PitcherId", "TaggedPitchType", "YEAR")

head(pitcherYearAvgs, 50)

pitcherYearAvgs$RelSpeedDiffYear <- pitcherYearAvgs$RelSpeed - pitcherYearAvgs$avg_RelSpeedYear
pitcherYearAvgs$VertBreakDiffYear <- pitcherYearAvgs$InducedVertBreak - pitcherYearAvgs$avg_IndVertBrkYear
pitcherYearAvgs$HorzBreakDiffYear <- pitcherYearAvgs$Hbrk - pitcherYearAvgs$avg_HBrkYear
pitcherYearAvgs$SpinRateDiffYear <- pitcherYearAvgs$SpinRate - pitcherYearAvgs$avg_spinrateYear

Year avgs would work, and might be better, for WHIP regressions and other results based tests.

HAve to make individual dfs for each pitch type

fastballsYear <- pitcherYearAvgs[pitcherYearAvgs$TaggedPitchType == 'Fastball', ]
changeupsYear <- pitcherYearAvgs[pitcherYearAvgs$TaggedPitchType == 'ChangeUp', ]
curveballsYear <- pitcherYearAvgs[pitcherYearAvgs$TaggedPitchType == 'Curveball', ]
cuttersYear <- pitcherYearAvgs[pitcherYearAvgs$TaggedPitchType == 'Cutter', ]
slidersYear <- pitcherYearAvgs[pitcherYearAvgs$TaggedPitchType == 'Slider', ]
sinkersYear <- pitcherYearAvgs[pitcherYearAvgs$TaggedPitchType == 'Sinker', ]
splittersYear <- pitcherYearAvgs[pitcherYearAvgs$TaggedPitchType == 'Splitter', ]

##Regressions

###Continuous

Fastballs

fastballSpinRateYear <- lm(SpinRateDiffYear ~ pitch_count, data = fastballsYear)
summary(fastballSpinRateYear)

## 
## Call:
## lm(formula = SpinRateDiffYear ~ pitch_count, data = fastballsYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1442.50   -56.15     4.81    62.88  1540.86 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  5.241863   0.192084   27.29   <2e-16 ***
## pitch_count -0.191874   0.005283  -36.32   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 100.9 on 634095 degrees of freedom
##   (3401 observations deleted due to missingness)
## Multiple R-squared:  0.002076,   Adjusted R-squared:  0.002074 
## F-statistic:  1319 on 1 and 634095 DF,  p-value: < 2.2e-16

fastballRelSpeedYear <- lm(RelSpeedDiffYear ~ pitch_count, data = fastballsYear)
summary(fastballRelSpeedYear)

## 
## Call:
## lm(formula = RelSpeedDiffYear ~ pitch_count, data = fastballsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -39.933  -0.768   0.072   0.881  14.643 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  0.3544947  0.0027376   129.5   <2e-16 ***
## pitch_count -0.0129755  0.0000753  -172.3   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.439 on 635019 degrees of freedom
##   (2477 observations deleted due to missingness)
## Multiple R-squared:  0.04468,    Adjusted R-squared:  0.04467 
## F-statistic: 2.97e+04 on 1 and 635019 DF,  p-value: < 2.2e-16

fastballVertBreakYear <- lm(VertBreakDiffYear ~ pitch_count, data = fastballsYear)
summary(fastballVertBreakYear)

## 
## Call:
## lm(formula = VertBreakDiffYear ~ pitch_count, data = fastballsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -41.517  -1.766   0.073   1.869  33.974 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  0.0438247  0.0058778   7.456 8.93e-14 ***
## pitch_count -0.0016038  0.0001616  -9.922  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.085 on 632796 degrees of freedom
##   (4700 observations deleted due to missingness)
## Multiple R-squared:  0.0001555,  Adjusted R-squared:  0.000154 
## F-statistic: 98.44 on 1 and 632796 DF,  p-value: < 2.2e-16

fastballHorzBreakYear <- lm(HorzBreakDiffYear ~ pitch_count, data = fastballsYear)
summary(fastballHorzBreakYear)

## 
## Call:
## lm(formula = HorzBreakDiffYear ~ pitch_count, data = fastballsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -44.538  -9.443  -0.854   9.534  37.277 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.0256665  0.0214953  -1.194    0.232
## pitch_count  0.0009401  0.0005915   1.589    0.112
## 
## Residual standard error: 11.25 on 629345 degrees of freedom
##   (8151 observations deleted due to missingness)
## Multiple R-squared:  4.014e-06,  Adjusted R-squared:  2.425e-06 
## F-statistic: 2.526 on 1 and 629345 DF,  p-value: 0.112

Change Up

changeupSpinRateYear <- lm(SpinRateDiffYear ~ pitch_count, data = changeupsYear)
summary(changeupSpinRateYear)

## 
## Call:
## lm(formula = SpinRateDiffYear ~ pitch_count, data = changeupsYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1150.81   -83.80    -6.27    70.90  2408.49 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  11.6678     0.7466   15.63   <2e-16 ***
## pitch_count  -0.3628     0.0183  -19.82   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 161.2 on 123178 degrees of freedom
##   (619 observations deleted due to missingness)
## Multiple R-squared:  0.003181,   Adjusted R-squared:  0.003172 
## F-statistic:   393 on 1 and 123178 DF,  p-value: < 2.2e-16

changeupRelSpeedYear <- lm(RelSpeedDiffYear ~ pitch_count, data = changeupsYear)
summary(changeupRelSpeedYear)

## 
## Call:
## lm(formula = RelSpeedDiffYear ~ pitch_count, data = changeupsYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -22.6986  -0.8695   0.0089   0.8904  12.9319 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  0.2661535  0.0072789   36.56   <2e-16 ***
## pitch_count -0.0082802  0.0001785  -46.39   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.575 on 123591 degrees of freedom
##   (206 observations deleted due to missingness)
## Multiple R-squared:  0.01712,    Adjusted R-squared:  0.01711 
## F-statistic:  2152 on 1 and 123591 DF,  p-value: < 2.2e-16

changeupVertBreakYear <- lm(VertBreakDiffYear ~ pitch_count, data = changeupsYear)
summary(changeupVertBreakYear)

## 
## Call:
## lm(formula = VertBreakDiffYear ~ pitch_count, data = changeupsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -35.704  -2.196   0.033   2.283  21.938 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept) -0.0979812  0.0179679  -5.453 4.96e-08 ***
## pitch_count  0.0030479  0.0004406   6.918 4.59e-12 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.881 on 123219 degrees of freedom
##   (578 observations deleted due to missingness)
## Multiple R-squared:  0.0003883,  Adjusted R-squared:  0.0003802 
## F-statistic: 47.86 on 1 and 123219 DF,  p-value: 4.591e-12

changeupHorzBreakYear <- lm(HorzBreakDiffYear ~ pitch_count, data = changeupsYear)
summary(changeupHorzBreakYear)

## 
## Call:
## lm(formula = HorzBreakDiffYear ~ pitch_count, data = changeupsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -33.727  -9.009  -0.349   9.099  33.723 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.035870   0.051072  -0.702    0.482
## pitch_count  0.001116   0.001252   0.891    0.373
## 
## Residual standard error: 10.99 on 122247 degrees of freedom
##   (1550 observations deleted due to missingness)
## Multiple R-squared:  6.496e-06,  Adjusted R-squared:  -1.684e-06 
## F-statistic: 0.7942 on 1 and 122247 DF,  p-value: 0.3728

Curveballs

curveballSpinRateYear <- lm(SpinRateDiffYear ~ pitch_count, data = curveballsYear)
summary(curveballSpinRateYear)

## 
## Call:
## lm(formula = SpinRateDiffYear ~ pitch_count, data = curveballsYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1560.49   -58.13     6.70    70.89  1955.26 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  5.59784    0.71775   7.799  6.3e-15 ***
## pitch_count -0.18349    0.01817 -10.097  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 134.9 on 87522 degrees of freedom
##   (1013 observations deleted due to missingness)
## Multiple R-squared:  0.001164,   Adjusted R-squared:  0.001152 
## F-statistic:   102 on 1 and 87522 DF,  p-value: < 2.2e-16

curveballRelSpeedYear <- lm(RelSpeedDiffYear ~ pitch_count, data = curveballsYear)
summary(curveballRelSpeedYear)

## 
## Call:
## lm(formula = RelSpeedDiffYear ~ pitch_count, data = curveballsYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -10.1420  -0.9747  -0.0423   0.9135  19.9656 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  0.1894395  0.0087500   21.65   <2e-16 ***
## pitch_count -0.0062093  0.0002215  -28.03   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.651 on 88291 degrees of freedom
##   (244 observations deleted due to missingness)
## Multiple R-squared:  0.008819,   Adjusted R-squared:  0.008808 
## F-statistic: 785.5 on 1 and 88291 DF,  p-value: < 2.2e-16

curveballVertBreakYear <- lm(VertBreakDiffYear ~ pitch_count, data = curveballsYear)
summary(curveballVertBreakYear)

## 
## Call:
## lm(formula = VertBreakDiffYear ~ pitch_count, data = curveballsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -24.273  -2.232  -0.176   1.876  39.805 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept) -0.0644487  0.0205406  -3.138   0.0017 ** 
## pitch_count  0.0021101  0.0005196   4.061 4.89e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.866 on 87887 degrees of freedom
##   (648 observations deleted due to missingness)
## Multiple R-squared:  0.0001876,  Adjusted R-squared:  0.0001762 
## F-statistic: 16.49 on 1 and 87887 DF,  p-value: 4.892e-05

curveballHorzBreakYear <- lm(HorzBreakDiffYear ~ pitch_count, data = curveballsYear)
summary(curveballHorzBreakYear)

## 
## Call:
## lm(formula = HorzBreakDiffYear ~ pitch_count, data = curveballsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -43.637  -8.962  -0.412   8.981  34.137 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.0052652  0.0582718  -0.090    0.928
## pitch_count  0.0001725  0.0014749   0.117    0.907
## 
## Residual standard error: 10.94 on 87453 degrees of freedom
##   (1082 observations deleted due to missingness)
## Multiple R-squared:  1.564e-07,  Adjusted R-squared:  -1.128e-05 
## F-statistic: 0.01368 on 1 and 87453 DF,  p-value: 0.9069

CUtters

cutterSpinRateYear <- lm(SpinRateDiffYear ~ pitch_count, data = cuttersYear)
summary(cutterSpinRateYear)

## 
## Call:
## lm(formula = SpinRateDiffYear ~ pitch_count, data = cuttersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1278.40   -53.47     3.33    61.35  1368.95 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  8.14612    1.55200   5.249 1.55e-07 ***
## pitch_count -0.27648    0.04013  -6.890 5.85e-12 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 115.9 on 13288 degrees of freedom
##   (110 observations deleted due to missingness)
## Multiple R-squared:  0.003559,   Adjusted R-squared:  0.003484 
## F-statistic: 47.47 on 1 and 13288 DF,  p-value: 5.848e-12

cutterlRelSpeedYear <- lm(RelSpeedDiffYear ~ pitch_count, data = cuttersYear)
summary(cutterlRelSpeedYear)

## 
## Call:
## lm(formula = RelSpeedDiffYear ~ pitch_count, data = cuttersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -13.2257  -0.8456   0.0077   0.8685  11.4163 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  0.2166739  0.0202787   10.69   <2e-16 ***
## pitch_count -0.0073515  0.0005242  -14.03   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.519 on 13371 degrees of freedom
##   (27 observations deleted due to missingness)
## Multiple R-squared:  0.0145, Adjusted R-squared:  0.01442 
## F-statistic: 196.7 on 1 and 13371 DF,  p-value: < 2.2e-16

cutterVertBreakYear <- lm(VertBreakDiffYear ~ pitch_count, data = cuttersYear)
summary(cutterVertBreakYear)

## 
## Call:
## lm(formula = VertBreakDiffYear ~ pitch_count, data = cuttersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -23.7201  -1.8558  -0.0654   1.7084  22.0430 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)  
## (Intercept) -0.083845   0.042984  -1.951   0.0511 .
## pitch_count  0.002844   0.001111   2.560   0.0105 *
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.217 on 13352 degrees of freedom
##   (46 observations deleted due to missingness)
## Multiple R-squared:  0.0004906,  Adjusted R-squared:  0.0004158 
## F-statistic: 6.554 on 1 and 13352 DF,  p-value: 0.01047

cutterHorzBreakYear <- lm(HorzBreakDiffYear ~ pitch_count, data = cuttersYear)
summary(cutterHorzBreakYear)

## 
## Call:
## lm(formula = HorzBreakDiffYear ~ pitch_count, data = cuttersYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -27.495  -8.992  -1.154   9.133  33.669 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)  
## (Intercept)  0.250328   0.147726   1.695   0.0902 .
## pitch_count -0.008522   0.003829  -2.225   0.0261 *
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 11.02 on 13242 degrees of freedom
##   (156 observations deleted due to missingness)
## Multiple R-squared:  0.0003738,  Adjusted R-squared:  0.0002983 
## F-statistic: 4.952 on 1 and 13242 DF,  p-value: 0.02608

Sliders

sliderSpinRateYear <- lm(SpinRateDiffYear ~ pitch_count, data = slidersYear)
summary(sliderSpinRateYear)

## 
## Call:
## lm(formula = SpinRateDiffYear ~ pitch_count, data = slidersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1916.90   -58.25    10.43    77.45  2294.62 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  6.30553    0.51832   12.16   <2e-16 ***
## pitch_count -0.21943    0.01377  -15.94   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 153.2 on 209264 degrees of freedom
##   (4771 observations deleted due to missingness)
## Multiple R-squared:  0.001212,   Adjusted R-squared:  0.001208 
## F-statistic:   254 on 1 and 209264 DF,  p-value: < 2.2e-16

sliderlRelSpeedYear <- lm(RelSpeedDiffYear ~ pitch_count, data = slidersYear)
summary(sliderlRelSpeedYear)

## 
## Call:
## lm(formula = RelSpeedDiffYear ~ pitch_count, data = slidersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -28.5727  -0.9945  -0.0015   0.9936  16.4665 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  0.2058581  0.0058695   35.07   <2e-16 ***
## pitch_count -0.0071664  0.0001559  -45.96   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.753 on 213530 degrees of freedom
##   (505 observations deleted due to missingness)
## Multiple R-squared:  0.009794,   Adjusted R-squared:  0.009789 
## F-statistic:  2112 on 1 and 213530 DF,  p-value: < 2.2e-16

sliderVertBreakYear <- lm(VertBreakDiffYear ~ pitch_count, data = slidersYear)
summary(sliderVertBreakYear)

## 
## Call:
## lm(formula = VertBreakDiffYear ~ pitch_count, data = slidersYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -27.222  -2.179  -0.039   2.082  56.100 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)  
## (Intercept) -0.0178076  0.0129526  -1.375   0.1692  
## pitch_count  0.0006200  0.0003442   1.801   0.0716 .
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.86 on 212707 degrees of freedom
##   (1328 observations deleted due to missingness)
## Multiple R-squared:  1.526e-05,  Adjusted R-squared:  1.056e-05 
## F-statistic: 3.245 on 1 and 212707 DF,  p-value: 0.07163

sliderHorzBreakYear <- lm(HorzBreakDiffYear ~ pitch_count, data = slidersYear)
summary(sliderHorzBreakYear)

## 
## Call:
## lm(formula = HorzBreakDiffYear ~ pitch_count, data = slidersYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -34.524  -9.244  -0.921   9.311  38.292 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.027955   0.037466  -0.746    0.456
## pitch_count  0.000975   0.000997   0.978    0.328
## 
## Residual standard error: 11.13 on 211266 degrees of freedom
##   (2769 observations deleted due to missingness)
## Multiple R-squared:  4.527e-06,  Adjusted R-squared:  -2.068e-07 
## F-statistic: 0.9563 on 1 and 211266 DF,  p-value: 0.3281

Sinkers

sliderSpinRateYear <- lm(SpinRateDiffYear ~ pitch_count, data = slidersYear)
summary(sliderSpinRateYear)

## 
## Call:
## lm(formula = SpinRateDiffYear ~ pitch_count, data = slidersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1916.90   -58.25    10.43    77.45  2294.62 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  6.30553    0.51832   12.16   <2e-16 ***
## pitch_count -0.21943    0.01377  -15.94   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 153.2 on 209264 degrees of freedom
##   (4771 observations deleted due to missingness)
## Multiple R-squared:  0.001212,   Adjusted R-squared:  0.001208 
## F-statistic:   254 on 1 and 209264 DF,  p-value: < 2.2e-16

sliderlRelSpeedYear <- lm(RelSpeedDiffYear ~ pitch_count, data = slidersYear)
summary(sliderlRelSpeedYear)

## 
## Call:
## lm(formula = RelSpeedDiffYear ~ pitch_count, data = slidersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -28.5727  -0.9945  -0.0015   0.9936  16.4665 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  0.2058581  0.0058695   35.07   <2e-16 ***
## pitch_count -0.0071664  0.0001559  -45.96   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.753 on 213530 degrees of freedom
##   (505 observations deleted due to missingness)
## Multiple R-squared:  0.009794,   Adjusted R-squared:  0.009789 
## F-statistic:  2112 on 1 and 213530 DF,  p-value: < 2.2e-16

sliderVertBreakYear <- lm(VertBreakDiffYear ~ pitch_count, data = slidersYear)
summary(sliderVertBreakYear)

## 
## Call:
## lm(formula = VertBreakDiffYear ~ pitch_count, data = slidersYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -27.222  -2.179  -0.039   2.082  56.100 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)  
## (Intercept) -0.0178076  0.0129526  -1.375   0.1692  
## pitch_count  0.0006200  0.0003442   1.801   0.0716 .
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.86 on 212707 degrees of freedom
##   (1328 observations deleted due to missingness)
## Multiple R-squared:  1.526e-05,  Adjusted R-squared:  1.056e-05 
## F-statistic: 3.245 on 1 and 212707 DF,  p-value: 0.07163

sliderHorzBreakYear <- lm(HorzBreakDiffYear ~ pitch_count, data = slidersYear)
summary(sliderHorzBreakYear)

## 
## Call:
## lm(formula = HorzBreakDiffYear ~ pitch_count, data = slidersYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -34.524  -9.244  -0.921   9.311  38.292 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.027955   0.037466  -0.746    0.456
## pitch_count  0.000975   0.000997   0.978    0.328
## 
## Residual standard error: 11.13 on 211266 degrees of freedom
##   (2769 observations deleted due to missingness)
## Multiple R-squared:  4.527e-06,  Adjusted R-squared:  -2.068e-07 
## F-statistic: 0.9563 on 1 and 211266 DF,  p-value: 0.3281

##Factor Level (Pitch_groups)

Fastballs

fastballSpinRateBinYear <- lm(SpinRateDiffYear ~ pitch_group, data = fastballsYear)
summary(fastballSpinRateBinYear)

## 
## Call:
## lm(formula = SpinRateDiffYear ~ pitch_group, data = fastballsYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1445.55   -56.12     4.81    62.86  1542.62 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                        2.7206     0.2404  11.317  < 2e-16 ***
## pitch_group10-19 Pitches           1.9217     0.3666   5.242 1.59e-07 ***
## pitch_group20-29 Pitches          -0.3470     0.4054  -0.856    0.392    
## pitch_group30-39 Pitches          -3.9807     0.4592  -8.669  < 2e-16 ***
## pitch_group40-49 Pitches          -6.1259     0.5215 -11.747  < 2e-16 ***
## pitch_group50-59 Pitches          -8.0382     0.5804 -13.850  < 2e-16 ***
## pitch_group60-69 Pitches         -12.3295     0.6343 -19.438  < 2e-16 ***
## pitch_group70-79 Pitches         -15.3186     0.7194 -21.295  < 2e-16 ***
## pitch_group80-89 Pitches         -15.7449     0.8620 -18.266  < 2e-16 ***
## pitch_group90-99 Pitches         -13.2271     1.1265 -11.742  < 2e-16 ***
## pitch_groupMore Than 100 Pitches -12.4073     1.6155  -7.680 1.59e-14 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 100.9 on 634086 degrees of freedom
##   (3401 observations deleted due to missingness)
## Multiple R-squared:  0.002544,   Adjusted R-squared:  0.002528 
## F-statistic: 161.7 on 10 and 634086 DF,  p-value: < 2.2e-16

#fastballSR <- as.data.frame(fastballSpinRateBinYear$coefficients)

fastballRelSpeedBinYear <- lm(RelSpeedDiffYear ~ pitch_group, data = fastballsYear)
summary(fastballRelSpeedBinYear)

## 
## Call:
## lm(formula = RelSpeedDiffYear ~ pitch_group, data = fastballsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -39.843  -0.766   0.074   0.880  14.762 
## 
## Coefficients:
##                                   Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                       0.361035   0.003423  105.46   <2e-16 ***
## pitch_group10-19 Pitches         -0.193071   0.005220  -36.98   <2e-16 ***
## pitch_group20-29 Pitches         -0.382668   0.005773  -66.28   <2e-16 ***
## pitch_group30-39 Pitches         -0.527655   0.006540  -80.69   <2e-16 ***
## pitch_group40-49 Pitches         -0.641584   0.007426  -86.40   <2e-16 ***
## pitch_group50-59 Pitches         -0.732228   0.008264  -88.60   <2e-16 ***
## pitch_group60-69 Pitches         -0.884686   0.009033  -97.94   <2e-16 ***
## pitch_group70-79 Pitches         -0.955556   0.010243  -93.29   <2e-16 ***
## pitch_group80-89 Pitches         -0.989537   0.012275  -80.62   <2e-16 ***
## pitch_group90-99 Pitches         -1.021188   0.016038  -63.67   <2e-16 ***
## pitch_groupMore Than 100 Pitches -0.953133   0.022994  -41.45   <2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.438 on 635010 degrees of freedom
##   (2477 observations deleted due to missingness)
## Multiple R-squared:  0.04671,    Adjusted R-squared:  0.04669 
## F-statistic:  3111 on 10 and 635010 DF,  p-value: < 2.2e-16

#as.data.frame(fastballRelSpeedBinYear$coefficients)

fastballVertBreakBinYear <- lm(VertBreakDiffYear ~ pitch_group, data = fastballsYear)
summary(fastballVertBreakBinYear)

## 
## Call:
## lm(formula = VertBreakDiffYear ~ pitch_group, data = fastballsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -41.507  -1.766   0.073   1.870  33.978 
## 
## Coefficients:
##                                   Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                       0.030503   0.007358   4.146 3.39e-05 ***
## pitch_group10-19 Pitches         -0.012767   0.011221  -1.138 0.255216    
## pitch_group20-29 Pitches         -0.008705   0.012408  -0.702 0.482983    
## pitch_group30-39 Pitches         -0.052571   0.014054  -3.741 0.000184 ***
## pitch_group40-49 Pitches         -0.054728   0.015958  -3.429 0.000605 ***
## pitch_group50-59 Pitches         -0.047091   0.017759  -2.652 0.008009 ** 
## pitch_group60-69 Pitches         -0.070036   0.019409  -3.608 0.000308 ***
## pitch_group70-79 Pitches         -0.114292   0.022010  -5.193 2.07e-07 ***
## pitch_group80-89 Pitches         -0.118241   0.026363  -4.485 7.29e-06 ***
## pitch_group90-99 Pitches         -0.230369   0.034469  -6.683 2.34e-11 ***
## pitch_groupMore Than 100 Pitches -0.174715   0.049467  -3.532 0.000413 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.085 on 632787 degrees of freedom
##   (4700 observations deleted due to missingness)
## Multiple R-squared:  0.0001751,  Adjusted R-squared:  0.0001593 
## F-statistic: 11.08 on 10 and 632787 DF,  p-value: < 2.2e-16

#fastballV <- as.data.frame(fastballVertBreakBinYear$coefficients)

fastballHorzBreakBinYear <- lm(HorzBreakDiffYear ~ pitch_group, data = fastballsYear)
summary(fastballHorzBreakBinYear)

## 
## Call:
## lm(formula = HorzBreakDiffYear ~ pitch_group, data = fastballsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -44.509  -9.444  -0.856   9.532  37.276 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)  
## (Intercept)                      -0.04607    0.02690  -1.713   0.0867 .
## pitch_group10-19 Pitches          0.03736    0.04104   0.910   0.3626  
## pitch_group20-29 Pitches          0.08406    0.04539   1.852   0.0640 .
## pitch_group30-39 Pitches          0.05912    0.05142   1.150   0.2502  
## pitch_group40-49 Pitches          0.06343    0.05837   1.087   0.2772  
## pitch_group50-59 Pitches          0.13665    0.06498   2.103   0.0355 *
## pitch_group60-69 Pitches          0.14970    0.07103   2.107   0.0351 *
## pitch_group70-79 Pitches         -0.08579    0.08058  -1.065   0.2870  
## pitch_group80-89 Pitches          0.06636    0.09653   0.687   0.4918  
## pitch_group90-99 Pitches          0.04705    0.12637   0.372   0.7096  
## pitch_groupMore Than 100 Pitches  0.09878    0.18080   0.546   0.5848  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 11.25 on 629336 degrees of freedom
##   (8151 observations deleted due to missingness)
## Multiple R-squared:  1.95e-05,   Adjusted R-squared:  3.609e-06 
## F-statistic: 1.227 on 10 and 629336 DF,  p-value: 0.2673

#fastballH <- as.data.frame(fastballHorzBreakBinYear$coefficients)

changeupSpinRateBinYear <- lm(SpinRateDiffYear ~ pitch_group, data = changeupsYear)
summary(changeupSpinRateBinYear)

## 
## Call:
## lm(formula = SpinRateDiffYear ~ pitch_group, data = changeupsYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1151.00   -83.82    -6.54    70.98  2409.04 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                         8.962      1.011   8.867  < 2e-16 ***
## pitch_group10-19 Pitches           -1.804      1.442  -1.251  0.21078    
## pitch_group20-29 Pitches           -5.467      1.578  -3.465  0.00053 ***
## pitch_group30-39 Pitches           -8.598      1.694  -5.075 3.89e-07 ***
## pitch_group40-49 Pitches          -14.517      1.791  -8.105 5.31e-16 ***
## pitch_group50-59 Pitches          -17.987      1.961  -9.173  < 2e-16 ***
## pitch_group60-69 Pitches          -18.682      2.143  -8.719  < 2e-16 ***
## pitch_group70-79 Pitches          -27.402      2.354 -11.640  < 2e-16 ***
## pitch_group80-89 Pitches          -27.361      2.707 -10.107  < 2e-16 ***
## pitch_group90-99 Pitches          -29.918      3.585  -8.346  < 2e-16 ***
## pitch_groupMore Than 100 Pitches  -33.496      5.335  -6.278 3.44e-10 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 161.2 on 123169 degrees of freedom
##   (619 observations deleted due to missingness)
## Multiple R-squared:  0.003179,   Adjusted R-squared:  0.003098 
## F-statistic: 39.29 on 10 and 123169 DF,  p-value: < 2.2e-16

#as.data.frame(changeupSpinRateBinYear$coefficients)

changeupRelSpeedBinYear <- lm(RelSpeedDiffYear ~ pitch_group, data = changeupsYear)
summary(changeupRelSpeedBinYear)

## 
## Call:
## lm(formula = RelSpeedDiffYear ~ pitch_group, data = changeupsYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -22.7409  -0.8689   0.0071   0.8889  12.8830 
## 
## Coefficients:
##                                   Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                       0.170935   0.009849  17.356  < 2e-16 ***
## pitch_group10-19 Pitches         -0.011529   0.014049  -0.821 0.411855    
## pitch_group20-29 Pitches         -0.052886   0.015376  -3.440 0.000583 ***
## pitch_group30-39 Pitches         -0.129105   0.016519  -7.815 5.52e-15 ***
## pitch_group40-49 Pitches         -0.254857   0.017464 -14.594  < 2e-16 ***
## pitch_group50-59 Pitches         -0.339838   0.019126 -17.769  < 2e-16 ***
## pitch_group60-69 Pitches         -0.494720   0.020887 -23.686  < 2e-16 ***
## pitch_group70-79 Pitches         -0.577324   0.022959 -25.146  < 2e-16 ***
## pitch_group80-89 Pitches         -0.645018   0.026410 -24.423  < 2e-16 ***
## pitch_group90-99 Pitches         -0.733432   0.034956 -20.981  < 2e-16 ***
## pitch_groupMore Than 100 Pitches -0.729123   0.051964 -14.031  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.574 on 123582 degrees of freedom
##   (206 observations deleted due to missingness)
## Multiple R-squared:  0.01806,    Adjusted R-squared:  0.01798 
## F-statistic: 227.3 on 10 and 123582 DF,  p-value: < 2.2e-16

#as.data.frame(changeupRelSpeedBinYear$coefficients)

changeupVertBreakBinYear <- lm(VertBreakDiffYear ~ pitch_group, data = changeupsYear)
summary(changeupVertBreakBinYear)

## 
## Call:
## lm(formula = VertBreakDiffYear ~ pitch_group, data = changeupsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -35.715  -2.195   0.030   2.282  21.855 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                      -0.10182    0.02432  -4.186 2.84e-05 ***
## pitch_group10-19 Pitches          0.02999    0.03469   0.865 0.387293    
## pitch_group20-29 Pitches          0.09407    0.03797   2.478 0.013227 *  
## pitch_group30-39 Pitches          0.14168    0.04080   3.473 0.000515 ***
## pitch_group40-49 Pitches          0.14364    0.04311   3.332 0.000863 ***
## pitch_group50-59 Pitches          0.26702    0.04725   5.651 1.60e-08 ***
## pitch_group60-69 Pitches          0.11880    0.05157   2.304 0.021247 *  
## pitch_group70-79 Pitches          0.30002    0.05667   5.294 1.20e-07 ***
## pitch_group80-89 Pitches          0.23491    0.06518   3.604 0.000314 ***
## pitch_group90-99 Pitches          0.21402    0.08629   2.480 0.013128 *  
## pitch_groupMore Than 100 Pitches -0.03948    0.12851  -0.307 0.758714    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.881 on 123210 degrees of freedom
##   (578 observations deleted due to missingness)
## Multiple R-squared:  0.0005564,  Adjusted R-squared:  0.0004752 
## F-statistic: 6.859 on 10 and 123210 DF,  p-value: 8.367e-11

#as.data.frame(changeupVertBreakBinYear$coefficients)

changeupHorzBreakBinYear <- lm(HorzBreakDiffYear ~ pitch_group, data = changeupsYear)
summary(changeupHorzBreakBinYear)

## 
## Call:
## lm(formula = HorzBreakDiffYear ~ pitch_group, data = changeupsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -33.890  -9.010  -0.349   9.083  33.684 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)  
## (Intercept)                      -0.01578    0.06912  -0.228    0.819  
## pitch_group10-19 Pitches         -0.07335    0.09861  -0.744    0.457  
## pitch_group20-29 Pitches          0.04296    0.10792   0.398    0.691  
## pitch_group30-39 Pitches          0.17918    0.11597   1.545    0.122  
## pitch_group40-49 Pitches          0.08605    0.12254   0.702    0.483  
## pitch_group50-59 Pitches         -0.14951    0.13440  -1.112    0.266  
## pitch_group60-69 Pitches         -0.05242    0.14670  -0.357    0.721  
## pitch_group70-79 Pitches         -0.08538    0.16098  -0.530    0.596  
## pitch_group80-89 Pitches          0.14784    0.18525   0.798    0.425  
## pitch_group90-99 Pitches          0.22619    0.24561   0.921    0.357  
## pitch_groupMore Than 100 Pitches  0.62245    0.36604   1.700    0.089 .
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 10.99 on 122238 degrees of freedom
##   (1550 observations deleted due to missingness)
## Multiple R-squared:  0.0001013,  Adjusted R-squared:  1.947e-05 
## F-statistic: 1.238 on 10 and 122238 DF,  p-value: 0.2604

#as.data.frame(changeupHorzBreakBinYear$coefficients)

Curveballs

curveballSpinRateBinYear <- lm(SpinRateDiffYear ~ pitch_group, data = curveballsYear)
summary(curveballSpinRateBinYear)

## 
## Call:
## lm(formula = SpinRateDiffYear ~ pitch_group, data = curveballsYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1562.89   -57.98     6.74    70.82  1951.94 
## 
## Coefficients:
##                                   Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                        1.13908    0.95270   1.196 0.231840    
## pitch_group10-19 Pitches           5.02238    1.37771   3.645 0.000267 ***
## pitch_group20-29 Pitches           1.92152    1.51975   1.264 0.206102    
## pitch_group30-39 Pitches           0.01574    1.67678   0.009 0.992510    
## pitch_group40-49 Pitches          -2.34638    1.83787  -1.277 0.201718    
## pitch_group50-59 Pitches          -8.46786    1.99723  -4.240 2.24e-05 ***
## pitch_group60-69 Pitches          -9.88622    2.19267  -4.509 6.53e-06 ***
## pitch_group70-79 Pitches         -10.46660    2.41838  -4.328 1.51e-05 ***
## pitch_group80-89 Pitches         -15.27392    2.76171  -5.531 3.20e-08 ***
## pitch_group90-99 Pitches          -9.76875    3.57655  -2.731 0.006309 ** 
## pitch_groupMore Than 100 Pitches -11.42645    4.93958  -2.313 0.020712 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 134.8 on 87513 degrees of freedom
##   (1013 observations deleted due to missingness)
## Multiple R-squared:  0.00163,    Adjusted R-squared:  0.001516 
## F-statistic: 14.29 on 10 and 87513 DF,  p-value: < 2.2e-16

#as.data.frame(curveballSpinRateBinYear$coefficients)

curveballRelSpeedBinYear <- lm(RelSpeedDiffYear ~ pitch_group, data = curveballsYear)
summary(curveballRelSpeedBinYear)

## 
## Call:
## lm(formula = RelSpeedDiffYear ~ pitch_group, data = curveballsYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -10.0839  -0.9741  -0.0455   0.9101  20.0234 
## 
## Coefficients:
##                                   Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                       0.113009   0.011614   9.731  < 2e-16 ***
## pitch_group10-19 Pitches          0.004884   0.016792   0.291    0.771    
## pitch_group20-29 Pitches         -0.008058   0.018520  -0.435    0.663    
## pitch_group30-39 Pitches         -0.116465   0.020429  -5.701 1.19e-08 ***
## pitch_group40-49 Pitches         -0.155680   0.022403  -6.949 3.70e-12 ***
## pitch_group50-59 Pitches         -0.284222   0.024328 -11.683  < 2e-16 ***
## pitch_group60-69 Pitches         -0.380180   0.026727 -14.225  < 2e-16 ***
## pitch_group70-79 Pitches         -0.441171   0.029470 -14.970  < 2e-16 ***
## pitch_group80-89 Pitches         -0.528695   0.033677 -15.699  < 2e-16 ***
## pitch_group90-99 Pitches         -0.496301   0.043644 -11.372  < 2e-16 ***
## pitch_groupMore Than 100 Pitches -0.459348   0.060114  -7.641 2.17e-14 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.65 on 88282 degrees of freedom
##   (244 observations deleted due to missingness)
## Multiple R-squared:  0.009777,   Adjusted R-squared:  0.009665 
## F-statistic: 87.17 on 10 and 88282 DF,  p-value: < 2.2e-16

#as.data.frame(curveballRelSpeedBinYear$coefficients)

curveballVertBreakBinYear <- lm(VertBreakDiffYear ~ pitch_group, data = curveballsYear)
summary(curveballVertBreakBinYear)

## 
## Call:
## lm(formula = VertBreakDiffYear ~ pitch_group, data = curveballsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -24.314  -2.232  -0.176   1.873  39.778 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                      -0.11697    0.02729  -4.286 1.82e-05 ***
## pitch_group10-19 Pitches          0.11213    0.03944   2.843 0.004470 ** 
## pitch_group20-29 Pitches          0.14229    0.04350   3.271 0.001072 ** 
## pitch_group30-39 Pitches          0.16648    0.04798   3.470 0.000521 ***
## pitch_group40-49 Pitches          0.18215    0.05258   3.464 0.000532 ***
## pitch_group50-59 Pitches          0.16197    0.05710   2.837 0.004558 ** 
## pitch_group60-69 Pitches          0.19078    0.06274   3.041 0.002360 ** 
## pitch_group70-79 Pitches          0.18386    0.06910   2.661 0.007801 ** 
## pitch_group80-89 Pitches          0.18032    0.07900   2.282 0.022464 *  
## pitch_group90-99 Pitches          0.22444    0.10231   2.194 0.028260 *  
## pitch_groupMore Than 100 Pitches  0.05407    0.14082   0.384 0.701010    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.866 on 87878 degrees of freedom
##   (648 observations deleted due to missingness)
## Multiple R-squared:  0.0003229,  Adjusted R-squared:  0.0002091 
## F-statistic: 2.838 on 10 and 87878 DF,  p-value: 0.00157

#as.data.frame(curveballVertBreakBinYear$coefficients)


curveballHorzBreakBinYear <- lm(HorzBreakDiffYear ~ pitch_group, data = curveballsYear)
summary(curveballHorzBreakBinYear)

## 
## Call:
## lm(formula = HorzBreakDiffYear ~ pitch_group, data = curveballsYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -43.697  -8.961  -0.419   8.985  34.121 
## 
## Coefficients:
##                                   Estimate Std. Error t value Pr(>|t|)
## (Intercept)                       0.001758   0.077378   0.023    0.982
## pitch_group10-19 Pitches         -0.043857   0.111885  -0.392    0.695
## pitch_group20-29 Pitches          0.056363   0.123429   0.457    0.648
## pitch_group30-39 Pitches          0.015103   0.136132   0.111    0.912
## pitch_group40-49 Pitches         -0.118843   0.149226  -0.796    0.426
## pitch_group50-59 Pitches          0.123086   0.161908   0.760    0.447
## pitch_group60-69 Pitches         -0.175070   0.178197  -0.982    0.326
## pitch_group70-79 Pitches          0.072103   0.196116   0.368    0.713
## pitch_group80-89 Pitches          0.145017   0.224120   0.647    0.518
## pitch_group90-99 Pitches          0.049306   0.291437   0.169    0.866
## pitch_groupMore Than 100 Pitches  0.002830   0.399660   0.007    0.994
## 
## Residual standard error: 10.94 on 87444 degrees of freedom
##   (1082 observations deleted due to missingness)
## Multiple R-squared:  4.718e-05,  Adjusted R-squared:  -6.717e-05 
## F-statistic: 0.4126 on 10 and 87444 DF,  p-value: 0.9415

#as.data.frame(curveballHorzBreakBinYear$coefficients)

Cutters

cutterSpinRateBinYear <- lm(SpinRateDiffYear ~ pitch_group, data = cuttersYear)
summary(cutterSpinRateBinYear)

## 
## Call:
## lm(formula = SpinRateDiffYear ~ pitch_group, data = cuttersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1276.82   -53.80     3.38    61.36  1365.15 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                        5.7401     2.0225   2.838 0.004544 ** 
## pitch_group10-19 Pitches          -2.4351     2.9862  -0.815 0.414828    
## pitch_group20-29 Pitches           0.1838     3.2922   0.056 0.955469    
## pitch_group30-39 Pitches          -3.7480     3.6487  -1.027 0.304344    
## pitch_group40-49 Pitches         -10.3194     4.1234  -2.503 0.012339 *  
## pitch_group50-59 Pitches         -20.2399     4.4892  -4.509 6.58e-06 ***
## pitch_group60-69 Pitches         -17.8565     5.0607  -3.528 0.000419 ***
## pitch_group70-79 Pitches         -22.8073     5.4923  -4.153 3.31e-05 ***
## pitch_group80-89 Pitches         -12.5231     6.2138  -2.015 0.043885 *  
## pitch_group90-99 Pitches         -26.5169     7.5197  -3.526 0.000423 ***
## pitch_groupMore Than 100 Pitches -18.6500    10.8137  -1.725 0.084610 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 115.9 on 13279 degrees of freedom
##   (110 observations deleted due to missingness)
## Multiple R-squared:  0.004475,   Adjusted R-squared:  0.003726 
## F-statistic: 5.969 on 10 and 13279 DF,  p-value: 4.355e-09

#as.data.frame(cutterSpinRateBinYear$coefficient)

cutterRelSpeedBinYear <- lm(RelSpeedDiffYear ~ pitch_group, data = cuttersYear)
summary(cutterRelSpeedBinYear)

## 
## Call:
## lm(formula = RelSpeedDiffYear ~ pitch_group, data = cuttersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -13.2597  -0.8471   0.0108   0.8661  11.4139 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                       0.20652    0.02642   7.818 5.79e-15 ***
## pitch_group10-19 Pitches         -0.10157    0.03901  -2.603  0.00924 ** 
## pitch_group20-29 Pitches         -0.17158    0.04301  -3.989 6.66e-05 ***
## pitch_group30-39 Pitches         -0.26536    0.04770  -5.563 2.70e-08 ***
## pitch_group40-49 Pitches         -0.33290    0.05386  -6.181 6.56e-10 ***
## pitch_group50-59 Pitches         -0.41626    0.05865  -7.097 1.34e-12 ***
## pitch_group60-69 Pitches         -0.39435    0.06611  -5.965 2.51e-09 ***
## pitch_group70-79 Pitches         -0.60866    0.07165  -8.495  < 2e-16 ***
## pitch_group80-89 Pitches         -0.68727    0.08113  -8.472  < 2e-16 ***
## pitch_group90-99 Pitches         -0.67949    0.09799  -6.934 4.28e-12 ***
## pitch_groupMore Than 100 Pitches -0.31865    0.14170  -2.249  0.02455 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.519 on 13362 degrees of freedom
##   (27 observations deleted due to missingness)
## Multiple R-squared:  0.01563,    Adjusted R-squared:  0.01489 
## F-statistic: 21.21 on 10 and 13362 DF,  p-value: < 2.2e-16

#as.data.frame(cutterRelSpeedBinYear$coefficient)

cutterVertBreakBinYear <- lm(VertBreakDiffYear ~ pitch_group, data = cuttersYear)
summary(cutterVertBreakBinYear)

## 
## Call:
## lm(formula = VertBreakDiffYear ~ pitch_group, data = cuttersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -23.7816  -1.8442  -0.0694   1.7192  21.9358 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)  
## (Intercept)                      -0.03930    0.05602  -0.702   0.4830  
## pitch_group10-19 Pitches         -0.01649    0.08270  -0.199   0.8420  
## pitch_group20-29 Pitches         -0.07245    0.09122  -0.794   0.4271  
## pitch_group30-39 Pitches          0.02628    0.10113   0.260   0.7950  
## pitch_group40-49 Pitches          0.19637    0.11416   1.720   0.0854 .
## pitch_group50-59 Pitches          0.23781    0.12427   1.914   0.0557 .
## pitch_group60-69 Pitches          0.20182    0.14006   1.441   0.1496  
## pitch_group70-79 Pitches         -0.03363    0.15179  -0.222   0.8246  
## pitch_group80-89 Pitches          0.01865    0.17187   0.109   0.9136  
## pitch_group90-99 Pitches          0.50875    0.20796   2.446   0.0144 *
## pitch_groupMore Than 100 Pitches  0.20207    0.30017   0.673   0.5008  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.217 on 13343 degrees of freedom
##   (46 observations deleted due to missingness)
## Multiple R-squared:  0.001284,   Adjusted R-squared:  0.0005351 
## F-statistic: 1.715 on 10 and 13343 DF,  p-value: 0.07127

#as.data.frame(cutterVertBreakBinYear$coefficient)

cutterHorzBreakBinYear <- lm(HorzBreakDiffYear ~ pitch_group, data = cuttersYear)
summary(cutterHorzBreakBinYear)

## 
## Call:
## lm(formula = HorzBreakDiffYear ~ pitch_group, data = cuttersYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -27.727  -8.952  -1.156   9.130  33.587 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)   
## (Intercept)                       0.11248    0.19224   0.585  0.55849   
## pitch_group10-19 Pitches         -0.15201    0.28406  -0.535  0.59255   
## pitch_group20-29 Pitches          0.01521    0.31332   0.049  0.96129   
## pitch_group30-39 Pitches          0.22605    0.34802   0.650  0.51601   
## pitch_group40-49 Pitches         -0.03972    0.39237  -0.101  0.91937   
## pitch_group50-59 Pitches         -0.04007    0.42810  -0.094  0.92542   
## pitch_group60-69 Pitches         -0.46202    0.48122  -0.960  0.33702   
## pitch_group70-79 Pitches         -0.13910    0.52492  -0.265  0.79102   
## pitch_group80-89 Pitches         -1.56235    0.59431  -2.629  0.00858 **
## pitch_group90-99 Pitches         -0.81362    0.72166  -1.127  0.25958   
## pitch_groupMore Than 100 Pitches -1.57748    1.03253  -1.528  0.12659   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 11.02 on 13233 degrees of freedom
##   (156 observations deleted due to missingness)
## Multiple R-squared:  0.0009412,  Adjusted R-squared:  0.0001862 
## F-statistic: 1.247 on 10 and 13233 DF,  p-value: 0.2552

#as.data.frame(cutterHorzBreakBinYear$coefficient)

Sliders

sliderSpinRateBinYear <- lm(SpinRateDiffYear ~ pitch_group, data = slidersYear)
summary(sliderSpinRateBinYear)

## 
## Call:
## lm(formula = SpinRateDiffYear ~ pitch_group, data = slidersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -1919.61   -58.25    10.39    77.50  2297.09 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                        3.1797     0.6708   4.740 2.14e-06 ***
## pitch_group10-19 Pitches           2.3261     0.9830   2.366   0.0180 *  
## pitch_group20-29 Pitches           0.9530     1.0943   0.871   0.3838    
## pitch_group30-39 Pitches          -5.2996     1.2216  -4.338 1.44e-05 ***
## pitch_group40-49 Pitches          -8.2343     1.3638  -6.038 1.56e-09 ***
## pitch_group50-59 Pitches         -10.7516     1.5018  -7.159 8.15e-13 ***
## pitch_group60-69 Pitches         -12.8987     1.6581  -7.779 7.34e-15 ***
## pitch_group70-79 Pitches         -15.9171     1.8620  -8.548  < 2e-16 ***
## pitch_group80-89 Pitches         -13.7920     2.1527  -6.407 1.49e-10 ***
## pitch_group90-99 Pitches         -19.6681     2.8346  -6.939 3.97e-12 ***
## pitch_groupMore Than 100 Pitches  -7.6781     4.0257  -1.907   0.0565 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 153.2 on 209255 degrees of freedom
##   (4771 observations deleted due to missingness)
## Multiple R-squared:  0.001513,   Adjusted R-squared:  0.001465 
## F-statistic: 31.71 on 10 and 209255 DF,  p-value: < 2.2e-16

#as.data.frame(sliderSpinRateBinYear$coefficient)

sliderRelSpeedBinYear <- lm(RelSpeedDiffYear ~ pitch_group, data = slidersYear)
summary(sliderRelSpeedBinYear)

## 
## Call:
## lm(formula = RelSpeedDiffYear ~ pitch_group, data = slidersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -28.5091  -0.9942  -0.0013   0.9925  16.4960 
## 
## Coefficients:
##                                   Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                       0.140549   0.007593  18.509  < 2e-16 ***
## pitch_group10-19 Pitches         -0.007899   0.011129  -0.710    0.478    
## pitch_group20-29 Pitches         -0.083878   0.012392  -6.769  1.3e-11 ***
## pitch_group30-39 Pitches         -0.154730   0.013829 -11.189  < 2e-16 ***
## pitch_group40-49 Pitches         -0.245708   0.015445 -15.908  < 2e-16 ***
## pitch_group50-59 Pitches         -0.302369   0.017004 -17.782  < 2e-16 ***
## pitch_group60-69 Pitches         -0.471239   0.018803 -25.062  < 2e-16 ***
## pitch_group70-79 Pitches         -0.517617   0.021109 -24.521  < 2e-16 ***
## pitch_group80-89 Pitches         -0.621009   0.024365 -25.487  < 2e-16 ***
## pitch_group90-99 Pitches         -0.605815   0.032147 -18.845  < 2e-16 ***
## pitch_groupMore Than 100 Pitches -0.506722   0.045360 -11.171  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.752 on 213521 degrees of freedom
##   (505 observations deleted due to missingness)
## Multiple R-squared:  0.01053,    Adjusted R-squared:  0.01049 
## F-statistic: 227.3 on 10 and 213521 DF,  p-value: < 2.2e-16

#as.data.frame(sliderRelSpeedBinYear$coefficient)

sliderVertBreakBinYear <- lm(VertBreakDiffYear ~ pitch_group, data = slidersYear)
summary(sliderVertBreakBinYear)

## 
## Call:
## lm(formula = VertBreakDiffYear ~ pitch_group, data = slidersYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -27.285  -2.178  -0.038   2.083  56.125 
## 
## Coefficients:
##                                    Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                      -0.0410648  0.0167616  -2.450 0.014289 *  
## pitch_group10-19 Pitches          0.0495680  0.0245674   2.018 0.043631 *  
## pitch_group20-29 Pitches          0.0269975  0.0273479   0.987 0.323553    
## pitch_group30-39 Pitches          0.1084424  0.0305254   3.553 0.000382 ***
## pitch_group40-49 Pitches          0.0484361  0.0340953   1.421 0.155431    
## pitch_group50-59 Pitches          0.0322589  0.0375425   0.859 0.390197    
## pitch_group60-69 Pitches          0.0391902  0.0415190   0.944 0.345216    
## pitch_group70-79 Pitches          0.1816247  0.0466087   3.897 9.75e-05 ***
## pitch_group80-89 Pitches          0.0008113  0.0537635   0.015 0.987961    
## pitch_group90-99 Pitches         -0.0380980  0.0709716  -0.537 0.591402    
## pitch_groupMore Than 100 Pitches  0.0889536  0.1002269   0.888 0.374798    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 3.86 on 212698 degrees of freedom
##   (1328 observations deleted due to missingness)
## Multiple R-squared:  0.000124,   Adjusted R-squared:  7.699e-05 
## F-statistic: 2.638 on 10 and 212698 DF,  p-value: 0.003266

#as.data.frame(sliderVertBreakBinYear$coefficient)


sliderHorzBreakBinYear <- lm(HorzBreakDiffYear ~ pitch_group, data = slidersYear)
summary(sliderHorzBreakBinYear)

## 
## Call:
## lm(formula = HorzBreakDiffYear ~ pitch_group, data = slidersYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -34.437  -9.242  -0.921   9.320  38.222 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)   
## (Intercept)                       0.05040    0.04844   1.041  0.29804   
## pitch_group10-19 Pitches         -0.19822    0.07104  -2.790  0.00526 **
## pitch_group20-29 Pitches         -0.01725    0.07911  -0.218  0.82739   
## pitch_group30-39 Pitches          0.02523    0.08832   0.286  0.77512   
## pitch_group40-49 Pitches         -0.11963    0.09860  -1.213  0.22499   
## pitch_group50-59 Pitches          0.02370    0.10860   0.218  0.82723   
## pitch_group60-69 Pitches         -0.07735    0.12032  -0.643  0.52031   
## pitch_group70-79 Pitches          0.04665    0.13505   0.345  0.72979   
## pitch_group80-89 Pitches          0.08510    0.15584   0.546  0.58502   
## pitch_group90-99 Pitches          0.09221    0.20635   0.447  0.65498   
## pitch_groupMore Than 100 Pitches -0.14450    0.29100  -0.497  0.61950   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 11.13 on 211257 degrees of freedom
##   (2769 observations deleted due to missingness)
## Multiple R-squared:  6.55e-05,   Adjusted R-squared:  1.817e-05 
## F-statistic: 1.384 on 10 and 211257 DF,  p-value: 0.1805

#as.data.frame(sliderHorzBreakBinYear$coefficient)

Sinkers

sinkersSpinRateBinYear <- lm(SpinRateDiffYear ~ pitch_group, data = sinkersYear)
summary(sinkersSpinRateBinYear)

## 
## Call:
## lm(formula = SpinRateDiffYear ~ pitch_group, data = sinkersYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -989.87  -52.80    1.22   56.76  710.06 
## 
## Coefficients:
##                                    Estimate Std. Error t value Pr(>|t|)   
## (Intercept)                        1.219078   1.407269   0.866  0.38635   
## pitch_group10-19 Pitches           4.794218   2.140123   2.240  0.02509 * 
## pitch_group20-29 Pitches          -0.360638   2.326097  -0.155  0.87679   
## pitch_group30-39 Pitches          -1.471929   2.671352  -0.551  0.58164   
## pitch_group40-49 Pitches          -4.527639   3.007504  -1.505  0.13223   
## pitch_group50-59 Pitches          -6.829206   3.340819  -2.044  0.04095 * 
## pitch_group60-69 Pitches         -11.422726   3.607824  -3.166  0.00155 **
## pitch_group70-79 Pitches         -12.984389   4.004572  -3.242  0.00119 **
## pitch_group80-89 Pitches          -0.003152   4.488724  -0.001  0.99944   
## pitch_group90-99 Pitches         -12.918962   5.539613  -2.332  0.01971 * 
## pitch_groupMore Than 100 Pitches   2.152943   8.189170   0.263  0.79263   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 97.48 on 17972 degrees of freedom
##   (62 observations deleted due to missingness)
## Multiple R-squared:  0.002452,   Adjusted R-squared:  0.001897 
## F-statistic: 4.417 on 10 and 17972 DF,  p-value: 3.124e-06

sinkersRelSpeedBinYear <- lm(RelSpeedDiffYear ~ pitch_group, data = sinkersYear)
summary(sinkersRelSpeedBinYear)

## 
## Call:
## lm(formula = RelSpeedDiffYear ~ pitch_group, data = sinkersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -15.9628  -0.6982   0.0325   0.7699   7.7123 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                       0.34539    0.01891  18.266  < 2e-16 ***
## pitch_group10-19 Pitches         -0.21119    0.02874  -7.349 2.08e-13 ***
## pitch_group20-29 Pitches         -0.28972    0.03125  -9.271  < 2e-16 ***
## pitch_group30-39 Pitches         -0.48009    0.03589 -13.376  < 2e-16 ***
## pitch_group40-49 Pitches         -0.58259    0.04039 -14.423  < 2e-16 ***
## pitch_group50-59 Pitches         -0.72496    0.04490 -16.147  < 2e-16 ***
## pitch_group60-69 Pitches         -0.83242    0.04846 -17.177  < 2e-16 ***
## pitch_group70-79 Pitches         -0.91636    0.05375 -17.049  < 2e-16 ***
## pitch_group80-89 Pitches         -0.85191    0.06027 -14.134  < 2e-16 ***
## pitch_group90-99 Pitches         -0.93975    0.07445 -12.623  < 2e-16 ***
## pitch_groupMore Than 100 Pitches -0.85333    0.11006  -7.754 9.41e-15 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.31 on 17993 degrees of freedom
##   (41 observations deleted due to missingness)
## Multiple R-squared:  0.04985,    Adjusted R-squared:  0.04932 
## F-statistic:  94.4 on 10 and 17993 DF,  p-value: < 2.2e-16

sinkersVertBreakBinYear <- lm(VertBreakDiffYear ~ pitch_group, data = sinkersYear)
summary(sinkersVertBreakBinYear)

## 
## Call:
## lm(formula = VertBreakDiffYear ~ pitch_group, data = sinkersYear)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -21.5546  -1.6612   0.0173   1.6575  17.4043 
## 
## Coefficients:
##                                   Estimate Std. Error t value Pr(>|t|)  
## (Intercept)                      -0.032401   0.040722  -0.796   0.4262  
## pitch_group10-19 Pitches          0.087538   0.061873   1.415   0.1571  
## pitch_group20-29 Pitches          0.058048   0.067317   0.862   0.3885  
## pitch_group30-39 Pitches         -0.041624   0.077289  -0.539   0.5902  
## pitch_group40-49 Pitches         -0.012714   0.086989  -0.146   0.8838  
## pitch_group50-59 Pitches         -0.008553   0.096734  -0.088   0.9295  
## pitch_group60-69 Pitches          0.070180   0.104330   0.673   0.5012  
## pitch_group70-79 Pitches         -0.055414   0.115709  -0.479   0.6320  
## pitch_group80-89 Pitches          0.094518   0.129866   0.728   0.4667  
## pitch_group90-99 Pitches          0.366821   0.160268   2.289   0.0221 *
## pitch_groupMore Than 100 Pitches  0.070593   0.236922   0.298   0.7657  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 2.82 on 17980 degrees of freedom
##   (54 observations deleted due to missingness)
## Multiple R-squared:  0.0005471,  Adjusted R-squared:  -8.771e-06 
## F-statistic: 0.9842 on 10 and 17980 DF,  p-value: 0.4545

sinkersHorzBreakBinYear <- lm(HorzBreakDiffYear ~ pitch_group, data = sinkersYear)
summary(sinkersHorzBreakBinYear)

## 
## Call:
## lm(formula = HorzBreakDiffYear ~ pitch_group, data = sinkersYear)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -29.581  -8.800  -0.734   8.828  41.458 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)  
## (Intercept)                       0.07817    0.15745   0.496   0.6196  
## pitch_group10-19 Pitches         -0.14002    0.23928  -0.585   0.5584  
## pitch_group20-29 Pitches         -0.13743    0.26034  -0.528   0.5976  
## pitch_group30-39 Pitches          0.01033    0.29893   0.035   0.9724  
## pitch_group40-49 Pitches         -0.25693    0.33676  -0.763   0.4455  
## pitch_group50-59 Pitches          0.23568    0.37500   0.628   0.5297  
## pitch_group60-69 Pitches         -0.54405    0.40475  -1.344   0.1789  
## pitch_group70-79 Pitches         -0.50730    0.45096  -1.125   0.2606  
## pitch_group80-89 Pitches         -0.36109    0.50530  -0.715   0.4749  
## pitch_group90-99 Pitches          0.85636    0.62559   1.369   0.1711  
## pitch_groupMore Than 100 Pitches  1.95723    0.91587   2.137   0.0326 *
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 10.86 on 17800 degrees of freedom
##   (234 observations deleted due to missingness)
## Multiple R-squared:  0.0006843,  Adjusted R-squared:  0.0001229 
## F-statistic: 1.219 on 10 and 17800 DF,  p-value: 0.2727

##Visualizing Regression Coefficients

library(jtools)
plot_summs(fastballSpinRateBinYear, changeupSpinRateBinYear, curveballSpinRateBinYear, cutterSpinRateBinYear, sliderSpinRateBinYear, model.names = c('Fastball', 'ChangeUp', 'Curveball', 'Cutter', 'Slider'), omit.coefs = c('(Intercept)', 'pitchBin9.+. 100+'))

## Registered S3 methods overwritten by 'broom':
##   method            from  
##   tidy.glht         jtools
##   tidy.summary.glht jtools

## Loading required namespace: broom.mixed

Release Speed Summary Graph

plot_summs(fastballRelSpeedBinYear, changeupRelSpeedBinYear, curveballRelSpeedBinYear, cutterRelSpeedBinYear, sliderRelSpeedBinYear, model.names = c('Fastball', 'ChangeUp', 'Curveball', 'Cutter', 'Slider'), omit.coefs = c('(Intercept)', 'pitchBin9.+. 100+'))

VErt Break

plot_summs(fastballVertBreakBinYear, changeupVertBreakBinYear, curveballVertBreakBinYear, cutterVertBreakBinYear, sliderVertBreakBinYear, model.names = c('Fastball', 'ChangeUp', 'Curveball', 'Cutter', 'Slider'), omit.coefs = c('(Intercept)', 'pitchBin9.+. 100+'))

Horz Break

plot_summs(fastballHorzBreakBinYear, changeupHorzBreakBinYear, curveballHorzBreakBinYear, cutterHorzBreakBinYear, sliderHorzBreakBinYear, model.names = c('Fastball', 'ChangeUp', 'Curveball', 'Cutter', 'Slider'), omit.coefs = c('(Intercept)', 'pitchBin9.+. 100+'))

Regressing against WHIP and FIP

data1ac <- sqldf("select 
               CASE WHEN PitchCall = 'InPlay' THEN TaggedHitType
               WHEN PitchCall = 'BallCalled' AND KorBB = 'Walk' THEN KorBB 
               WHEN PitchCall = 'StrikeSwinging' AND KorBB = 'Strikeout' THEN 'SwingingStikeout'
               WHEN PitchCall = 'StrikeCalled' AND KorBB = 'Strikeout' THEN 'StrikeoutLooking'
               ELSE PitchCall END AS Result
               FROM pitches" )
pitches['Result'] = data1ac['Result']

pitches$TopOrBottom <- pitches$Top.Bottom

intermediate_step <- sqldf("SELECT PitcherId, BatterTeam, GameID, PitcherTeam, Date, Inning, TopOrBottom, PAofInning, BatterId, Result, PlayResult, KorBB, pitch_count - PitchofPA + 1 AS CountatFirstPitch, PitchofPA from pitches ORDER BY GameID, BatterTeam, Inning, PAofInning ")


length(intermediate_step$BatterId)

## [1] 1144455

#Well run into problems with the occasional batter without a BAtter ID (NA)
#Therefore, I'm going to replace the NA batterIds it 0000 or something

#First lets see how many there are
mean(is.na(intermediate_step$BatterId))

## [1] 0.0002621335

#looks like .026 percent. Thats not bad. It isnt a problem unless theres 2 consecuive batters with no id

#I'm going to make NA BatterIds equal 999999
intermediate_step$BatterId <- intermediate_step$BatterId %>% replace(is.na(.), 999999)


last_pitches = intermediate_step %>%
  group_by(GameID, Inning, BatterId, PAofInning) %>%
  slice_tail(n=1) %>%
  arrange(GameID, Inning, PitcherTeam, PAofInning)


dim(last_pitches)

## [1] 300503     14

#Neew code chucnk just to take a look at last_pitches
head(last_pitches, 300)

# Add a new factor column to the dataframe for the pitch group
last_pitches$first_pitch_bin <- as.factor(ifelse(last_pitches$CountatFirstPitch < 100, (last_pitches$CountatFirstPitch) %/% 10 + 1, 11))

# Check the updated dataframe
head(last_pitches, 250)

#Let make a first_Pitch_group Column

last_pitches$first_pitch_group <- NA

last_pitches$first_pitch_group[last_pitches$first_pitch_bin == '1'] <- '0-9 pitches'
last_pitches$first_pitch_group[last_pitches$first_pitch_bin == '2'] <- '10-19 pitches'
last_pitches$first_pitch_group[last_pitches$first_pitch_bin == '3'] <- '20-29 pitches'
last_pitches$first_pitch_group[last_pitches$first_pitch_bin == '4'] <- '30-39 pitches'
last_pitches$first_pitch_group[last_pitches$first_pitch_bin == '5'] <- '40-49 pitches'
last_pitches$first_pitch_group[last_pitches$first_pitch_bin == '6'] <- '50-59 pitches'
last_pitches$first_pitch_group[last_pitches$first_pitch_bin == '7'] <- '60-69 pitches'
last_pitches$first_pitch_group[last_pitches$first_pitch_bin == '8'] <- '70-79 pitches'
last_pitches$first_pitch_group[last_pitches$first_pitch_bin == '9'] <- '80-89 pitches'
last_pitches$first_pitch_group[last_pitches$first_pitch_bin == '10'] <- '90-99 pitches'
last_pitches$first_pitch_group[last_pitches$first_pitch_bin == '11'] <- 'More Than 100 Pitches'

#MAke sure the order is correct. Really annoying if regression output isn't in ascending order
sqldf("SELECT first_pitch_group, count(*) from last_pitches GROUP BY first_pitch_group ORDER BY first_pitch_group")

last_pitches$result2 <- NA
last_pitches$result3 <- NA

last_pitches$result2 <- ifelse(!(is.na(last_pitches$PlayResult)) & last_pitches$PlayResult != 'Undefined', last_pitches$PlayResult, NA)

last_pitches$result2 <- ifelse(!(is.na(last_pitches$KorBB)) & last_pitches$KorBB != 'Undefined', last_pitches$Result, last_pitches$result2)

last_pitches$result2 <- ifelse(!(is.na(last_pitches$Result)) & last_pitches$Result == 'HitByPitch', last_pitches$Result, last_pitches$result2)

last_pitches$result2 <- ifelse(is.na(last_pitches$result2), "###ISSUE###", last_pitches$result2)

last_pitches$result3 <- ifelse(is.na(last_pitches$result2), last_pitches$Result, last_pitches$result3)

sqldf("SELECT result2, count(*) FROM last_pitches GROUP BY result2")

newdata <- last_pitches[(last_pitches$result2 %in% c('Double','Error','FieldersChoice', 'Fielderschoice', 'HitByPitch', 'HomeRun', 'Homerun', 'Out', 'Sacrifice', 'Single', 'StrikeoutLooking', 'SwingingStikeout', 'Triple', 'Walk')), ]

newdata1 <- newdata

newdata1$result2[newdata$result2 == 'Fielderschoice'] <- 'FieldersChoice'
newdata1$result2[newdata$result2 == 'Homerun'] <- 'HomeRun'

write.csv(newdata1, 'C:\\Users\\Nick\\UCSB Baseball\\NickAllCollegeTrackmanAtBatsNew.csv')

atBats <- read.csv('C:\\Users\\Nick\\UCSB Baseball\\NickAllCollegeTrackmanAtBatsNew.csv')

#Making some dummy columns
library(fastDummies)
lastPitches <- dummy_cols(atBats, select_columns = 'result2')

lastPitches$walksPlusHits <- lastPitches$result2_Walk + lastPitches$result2_Single + lastPitches$result2_Double + lastPitches$result2_Triple + lastPitches$result2_HomeRun + lastPitches$result2_HitByPitch
lastPitches$outs <- lastPitches$result2_StrikeoutLooking + lastPitches$result2_SwingingStikeout + lastPitches$result2_FieldersChoice + lastPitches$result2_Out + lastPitches$result2_Sacrifice

whip_bin_avgs <- sqldf("SELECT first_pitch_group, count(*),
                   cast((sum(cast(walksPlusHits as float)) / sum(cast(outs as float)) * 3.00) as float) AS whip FROM lastPitches group by first_pitch_group")


whip_bin_avgs

whip_bin_pitcher_avgs <- sqldf("SELECT first_pitch_group, PitcherId, count(*) as numAtBats,
                   cast((sum(cast(walksPlusHits as float)) / sum(cast(outs as float)) * 3.00) as float) AS whip_bin FROM lastPitches group by first_pitch_group, PitcherId ORDER BY count(*) DESC")



whip_bin_pitcher_avgs_ten_plus <- whip_bin_pitcher_avgs[whip_bin_pitcher_avgs$numAtBats >= 4, ]

whip_pitcher_avgs <- sqldf("SELECT PitcherId, count(*),
                   cast((sum(cast(walksPlusHits as float)) / sum(cast(outs as float)) * 3.00) as float) AS whip_avg FROM lastPitches group by PitcherId ORDER BY count(*) DESC")

whip_pitcher_avgs

whipdf <- left_join(whip_bin_pitcher_avgs_ten_plus, whip_pitcher_avgs, on = 'PitcherId')

## Joining, by = "PitcherId"

whipdf$diff <- whipdf$whip_bin - whipdf$whip_avg

whip_diff_mod <- lm(diff ~ first_pitch_group, data = whipdf, weights = numAtBats)
summary(whip_diff_mod)

## 
## Call:
## lm(formula = diff ~ first_pitch_group, data = whipdf, weights = numAtBats)
## 
## Weighted Residuals:
##     Min      1Q  Median      3Q     Max 
## -51.320  -2.796  -0.903   1.664  84.547 
## 
## Coefficients:
##                                        Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                             0.14448    0.01912   7.555 4.35e-14 ***
## first_pitch_group10-19 pitches          0.04997    0.03073   1.626 0.104020    
## first_pitch_group20-29 pitches          0.05781    0.03480   1.661 0.096650 .  
## first_pitch_group30-39 pitches          0.24717    0.03944   6.267 3.75e-10 ***
## first_pitch_group40-49 pitches          0.22669    0.04426   5.122 3.05e-07 ***
## first_pitch_group50-59 pitches          0.15657    0.04929   3.177 0.001493 ** 
## first_pitch_group60-69 pitches          0.25429    0.05391   4.716 2.42e-06 ***
## first_pitch_group70-79 pitches          0.24326    0.06176   3.939 8.22e-05 ***
## first_pitch_group80-89 pitches          0.20015    0.07473   2.678 0.007402 ** 
## first_pitch_group90-99 pitches          0.37981    0.10675   3.558 0.000375 ***
## first_pitch_groupMore Than 100 Pitches  0.29814    0.17317   1.722 0.085148 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 5.524 on 21553 degrees of freedom
##   (165 observations deleted due to missingness)
## Multiple R-squared:  0.004121,   Adjusted R-squared:  0.003659 
## F-statistic: 8.919 on 10 and 21553 DF,  p-value: 8.359e-15

whip_diff_mod

## 
## Call:
## lm(formula = diff ~ first_pitch_group, data = whipdf, weights = numAtBats)
## 
## Coefficients:
##                            (Intercept)          first_pitch_group10-19 pitches  
##                                0.14448                                 0.04997  
##         first_pitch_group20-29 pitches          first_pitch_group30-39 pitches  
##                                0.05781                                 0.24717  
##         first_pitch_group40-49 pitches          first_pitch_group50-59 pitches  
##                                0.22669                                 0.15657  
##         first_pitch_group60-69 pitches          first_pitch_group70-79 pitches  
##                                0.25429                                 0.24326  
##         first_pitch_group80-89 pitches          first_pitch_group90-99 pitches  
##                                0.20015                                 0.37981  
## first_pitch_groupMore Than 100 Pitches  
##                                0.29814

lastPitches$BBFIP <- (lastPitches$result2_Walk + lastPitches$result2_HitByPitch) * 3.00
lastPitches$HRFIP <- lastPitches$result2_HomeRun * 13.00
lastPitches$KFIP <- lastPitches$result2_Strikeout * (-2.00)
lastPitches$NUMERATOR <- (lastPitches$BBFIP + lastPitches$HRFIP + lastPitches$KFIP)*3

fip_bin_avgs <- sqldf("SELECT first_pitch_group, count(*),
                   sum(NUMERATOR)/sum(outs) + 3.90 AS fip FROM lastPitches group by first_pitch_group")


fip_bin_avgs